Вы находитесь на странице: 1из 43

DRAFT WHITE PAPER:

THE GLOBAL WATER CYCLE AND ITS ROLE IN


CLIMATE AND GLOBAL CHANGE
In support of Chapter 7 of the

Strategic Plan
for the
Climate Change Science Program
Draft dated 26 November 2002

US Climate Change Science Program


1717 Pennsylvania Ave., NW
Suite 250
Washington, DC 20006
Tel: +1 202 223 6262
Fax: +1 202 223 3065

DISCUSSION DRAFT

Authors and Contributors


Prepared by the USGCRP/ CCSP Interagency Working Group on the Global Water Cycle
Authors:
Rick Lawford
Susanna Eden
Wanda Ferrell
Harvey Hill
Jin Huang
Douglas James
Pam Stephens
Sushel Unninayar
Mike Dettinger
Jared Entin
David Goodrich
Bill Kirby
Mark Weltz
Reviewers:
Roni Avissar
George Hornberger
R. Csar Izaurralde
Dennis Lettenmaier
James Shuttleworth

DISCUSSION DRAFT

Preface
On 11 November 2002, the US Climate Change Science Program issued a discussion draft of its
Strategic Plan. The strategy for each major area of the program is summarized in specific

chapters of the draft plan, and for four chapters is described in greater detail in white
papers. The white papers, including this one focused on the water cycle, represent the
views of the authors and are not statements of policy or findings of the United States
Government or its Departments/Agencies. They are intended to support discussion
during the US Climate Change Science Program Planning Workshop for Scientists and
Stakeholders being held in Washington, DC on December 3 5, 2002.
Both the chapters of the plan and the white papers should be considered drafts.
Comments on the chapters of the draft Strategic Plan may be provided during the
USCCSP Planning Workshop on December 3 5, 2002, and during a subsequent public
comment period extending to January 13, 2003. The chapters of the Strategic Plan will
be subject to substantial revision based on these comments and on independent review by
the National Academy of Sciences. A final version of the Strategic Plan, setting a path
for the next few years of research under the CCSP, will be published by April 2003.
Information about the Workshop and opportunities for written comment is available on
the web site www.climatescience.gov.
Comments that are specific to this white paper and that are not already conveyed
through comments on the related chapter of the plan should be directed to: Susanna
Eden [seden@usgcrp.gov].

DISCUSSION DRAFT

1
2
3
4
5
6
7
8
9
10

DRAFT WHITE PAPER:


THE GLOBAL WATER CYCLE AND ITS ROLE IN
CLIMATE AND GLOBAL CHANGE
In support of Chapter 7 of the
Strategic Plan for the
Climate Change Science Program
Draft dated 25 November 2002

In this paper
1. Introduction
2. Elements of the CCSP Global Water Cycle Component
2.1. Internal Water Cycle Mechanisms
2.2. Water Cycle Feedback Effects on the Climate
2.3. Predicting Water Cycle Variability and Change
2.4. The Cycling of Water and other Biogeochemical Constituents
2.5. More Effective Water Management through Water Cycle Science
3. Summary
11

1. Introduction
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31

The Global Water Cycle (GWC) determines the amount of water that is available for
human uses such as municipal and industrial supply, irrigation and agriculture,
hydropower, waste disposal, protection of human and ecosystem health and a wide range
of societal and environmental benefits. The GWC is an integral part of the Earth/ Climate
system; water vapor constitutes the Earths most abundant and important greenhouse gas,
and water is its most active solvent. The interactions of the GWC and the climate system
manifest themselves through many processes and phenomena, such as cloud formation,
precipitation, groundwater recharge, accumulation and ablation of snow packs and
glaciers, droughts and floods. Furthermore, water regulates the Earths energy balance
because energy is absorbed (or released) when liquid water is converted to or from water
vapor and the energy stored in water is transferred from one location to another through
water transport. These properties account for the critical role that the cycling of water
plays in climate variability and its feedback effects that have a strong influence on the
rate of climate change. The Global Water Cycle program forms a distinctive element
within the CCSP that focuses research towards a more coherent view of the movements,
transformations, and reservoirs of water, energy and water-borne materials throughout the
Earth system and their interactions with ecosystems and human systems. In particular, the
GWC element contributes to climate science by providing research on critical areas of
uncertainty in climate change science and building the scientific basis needed by water

DISCUSSION DRAFT
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46

users and managers to adapt to climate variability and change in a sustainable way.
Although the GWC operates on a continuum of space and time scales, this document
places more emphasis on the time and spatial scales relevant to climate issues.
The water cycle is now widely recognized as one of the dominant causes of uncertainty in
climate change projections. Moreover, most major impacts of climate variability and
climate change on human activity and natural ecosystems directly involve precipitation
processes or water and energy cycles. Precipitation projections have been very uncertain;
climate models have produced contradictory projections for the central USAwhether it
will experience drying or wetting as atmospheric carbon dioxide concentrations increase.
Because the availability of water for human uses and ecosystem functions in this area (as
is the case in most land areas) is more sensitive to precipitation changes than changes in
temperature, these uncertainties are important to resolve. Furthermore, the responses of
vegetation and ecosystems to precipitation have implications for carbon sequestration.
Increases in carbon sequestration in northern hemisphere forests are being attributed to
changes in regional precipitation regimes. Furthermore, the changing evapotranspiration
rates associated with changing vegetative cover arising from land use change complicates
the interpretation of the warming trends that have been observed over the USA during the
past century.
The Earths water cycle is driven by processes that force the movement of water from
one reservoir to another. Evaporation from the oceans and land is the primary source of
atmospheric water vapor, which is transported, often over long distances, and eventually
condenses into cloud particles, that in turn develop into precipitation. Precipitation over
land finds its way into rivers, aquifers, and eventually oceans. Globally, there is as much
water precipitated as is evaporated, but over land precipitation exceeds evaporation and
over oceans evaporation exceeds precipitation. The excess precipitation over land equals
the flow of surface and groundwater from continents to the oceans. This natural cycling
of water is now perturbed by human activities. Together with changing vegetation
patterns due to land management practices, these factors complicate the prediction of the
consequences of climate change on the Global Water Cycle.
The water cycle is coupled with biogeochemical cycles that control the movement if
nutrients, waste products and even toxic chemicals, in aquatic-land and coastal
ecosystems. These linkages directly affect water quality and the availability of potable
water and industrial water supplies. Water supplies are subject to a range of stresses, such
as population growth, pollution and industrial and urban development. Furthermore,
water has been identified as a major factor in the occurrence and transmission of a
number of vector-borne diseases (e.g. West Nile virus). These issues lead to public
concerns about water quality and efforts to improve the management of fresh water
resources. Accordingly, the global water cycle is an issue of central concern in the USA
and in every other country of the world. The needs for adequate supplies of clean water
pose major challenges to social and economic development and to the management of
natural resources and ecosystems. These challenges grow ever greater as variations and
changes in climate alter the hydrologic cycle in ways that are currently unpredictable.

DISCUSSION DRAFT
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45

Extremes in the surface water cycle, in terms of droughts and floods, have major
implications for the security of life and property and for economic activity. Floods are
the most important natural hazard for the USA in terms of loss of life and property, with
annual average losses now approaching $10 billion per year. Although drought losses are
more difficult to quantify, the 1988 central US drought impacts have been estimated at
more than $20 billion.
A recent report to the federal government by a group of leading atmospheric and
hydrologic scientists, A Plan for a New Science Initiative on the Global Water Cycle.
(Hornberger et al., 2001) highlights the need for water cycle research. The authors
emphasized that the water cycle is changing in ways that we have never experienced, and
consequently, we must develop the knowledge base, information and decision support
resources needed to deal with these emerging realities. This report identified three major
areas that require more research over the next decade. The three areas include
documentation and understanding of trends and variability in the GWC, improving the
accuracy of water cycle predictive capabilities, and developing a better scientific
understanding of the linkages between the water cycle and other biogeochemical cycles,
especially the carbon cycle. These three areas form the basis for discussions in this
white paper related to climate change and water cycle trends, prediction and the linkages
between water and nutrients cycles in terrestrial and freshwater ecosystems.
The emerging capability to predict GWC variations at seasonal to interannual time scales
provides a basis for dialog between the scientific community and water system and land
managers. This dialog is enabling the program to provide the scientific underpinnings for
improving the adaptability of existing infrastructure and management practices. Potential
changes in water cycle variables such as precipitation, evaporation and runoff have
critical implications for agriculture, water supply and hydropower managers, and other
sectors that are affected by long-term water cycle changes.
To address the urgent need for better information on the water cycle, the USGCRP/CCSP
is planning its Global Water Cycle research program around two overarching questions,
namely:
1. How do water cycle processes (including climate feedbacks) and human activities
influence the distribution and quality of water within the Earth system, to what
extent are changes predictable, and how are these processes and activities linked
to ecosystem and human health and the cycling of important chemicals, such as
carbon, nitrogen, other nutrients, and toxic substances?
2. How will large-scale changes in climate, demographics, and land use (including
changes in agricultural and land management practices), affect the capacity of
societies to provide adequate supplies of clean water for human uses and
ecosystems and respond to extreme hydrologic events?
The above questions define the scope of a science-driven Water Cycle program focused
on the needs of society. To address these questions in a comprehensive way, the program
elements developed in this document deal with:

DISCUSSION DRAFT
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36

1) long term trends in the global water cycle and the internal water cycle processes
responsible for these trends;
2) links between the water cycle and the climate system and controls that the water
cycle places on climate variability and change;
3) development of a capability to predict water cycle variables;
4) linkages between variations in the water cycle and variations in other connected
cycles such as carbon and nitrogen and in reservoirs such as the cryosphere,
ecosystems and coastal areas; and
5) information on the impacts of water cycle variability and change and its use in
planning and management decisions that affect the use of the Nations water
resources.
USGCRP water cycle research, therefore, directly intersects all of the focused research
required by the CCSP, especially in the context of delivering scientific results, observing
and attributing trends and variability, assembling sets of integrated data and information,
and improving prediction products, as needed for the development of decision support
tools for water management.
Impacts of global change on water resources will be complicated and interactions will
involve feedbacks likely to produce surprises and unusual events. Advances in GWC
research require a mix of observational program enhancements, field experiments and
process studies, model development and testing, and modeling studies. The
development of better models deserves particular attention, as models are the key
building block for improving the accuracy of water cycle predictions. To address the
breadth of water cycle issues arising in the program, a mix of models is needed, some
within linked model hierarchies for prediction purposes and some that would be run
offline to build understanding of processes and provide the linkages with biogeochemical
cycling and decisions support. Models will be used in both simulation and prediction
modes. A suite of models and modeling strategies will be needed ranging from small
area process models operating in stand-alone fashion to regional models nested in a
hierarchy of partially- and fully-coupled models, to global Earth system models that
include the representation of all elements of the GWC. The following pages outline the
five questions that constitute the core of the CCSP water cycle element, as well as the
associated research needs and expected results.

2. Elements of the CCSP Global Water Cycle Component


37
38
39
40
41
42
43

2.1 INTERNAL WATER CYCLE MECHANISMS


QUESTION 1: WHAT ARE THE UNDERLYING MECHANISMS AND
PROCESSES RESPONSIBLE FOR THE MAINTENANCE AND
VARIABILITY OF THE WATER CYCLE; ARE THE
CHARACTERISTICS OF THE CYCLE CHANGING AND, IF SO, TO

DISCUSSION DRAFT
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45

WHAT EXTENT ARE HUMAN ACTIVITIES RESPONSIBLE FOR


THOSE CHANGES?
State of Knowledge
Water strongly influences the Earths radiation balance. Clouds reflect short wave
radiation (thereby cooling the atmosphere) and absorb terrestrial outgoing long-wave
radiation (warming the atmosphere). The nature and magnitude of the cooling/heating is
a function of the areal coverage, height, structure and optical properties of the clouds.
Water molecules are strong absorbers of infrared radiation and water vapor is, by far, the
most effective of the greenhouse gases. Water vapor concentrations in the upper
troposphere and lower stratosphere are very critical in determining the rate at which
radiative energy emitted by the atmosphere escapes to space.
The water cycle plays a key role in the maintenance of the climate system as a moderator
of the Earths energy cycle. It is through the water cycle that incoming solar energy is
redistributed through the Earth system via the atmosphere and oceans. Latent heat
exchanges occur as water changes phases from solid or liquid to vapor and vice versa.
Water cools its surroundings when it evaporates or sublimates, usually at the land or
ocean surface, and warms the surrounding air when vapor condenses as clouds and
precipitation. Water vapor is transported by the wind from its source to other regions.
Because of the large amount of heat released by the condensation of water molecules
(latent heat), water vapor is a very effective means of storing energy and the latent heat
flux in the atmosphere is a major component of the overall transport of energy from
equator to poles. Furthermore, latent heat is the principal source of energy that drives
cyclogenesis and sustains weather systems.
Recent observations suggest that there have been significant changes in a number of
water cycle components precipitation intensity, distribution and types; surface and
subsurface runoff; cloud cover properties; atmospheric water vapor; and river discharge.
For example, U.S., precipitation is characterized by more high intensity events than
occurred in the past. According to current climate model predictions, the most significant
manifestation of global warming could be an acceleration of the rate of the global water
cycle. However, current climate models do not make consistently accurate predictions,
and these critically important projections are very uncertain.
Within any part of the climate system, there is a substantial range of natural variability
due strictly to internal processes. It is extremely difficult to distinguish between the
natural excursions from the "norm" and changes that might be the result of forcing caused
by human actions. Major improvements in observations and models of the water cycle
are required in order to distinguish natural variability from change. Once models can
successfully simulate past water cycle behavior, they can be used to assess potential
changes due to human activity, such as anthropogenic emissions of greenhouse gases,
land use change or aerosol production.
Current models do not simulate many aspects of the global climate well, and many of the
model shortcomings are related to poor representations of the GWC. For a given

DISCUSSION DRAFT
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39

increase in CO2 , different climate models produce vastly different cloud, precipitation
and soil moisture (both in magnitude and sign) depending on their parameterizations of
basic water cycle processes. Furthermore, as global temperatures warm, the atmosphere
becomes capable of holding more moisture and the warmer temperatures normally
increase evaporation rates and the amount of water vapor throughout the atmosphere.
This basic knowledge, however, does not reveal whether the increased atmospheric water
vapor will lead to the formation of more extensive cloud covers and will enhance, reduce
or counteract global warming.

40
41
42
43
44
45

Draft Research Questions

Other external factors must also be considered in these models. Atmospheric aerosols
affect cloud condensation nuclei concentrations, the radiative properties of cloud
particles, and precipitation processes. Changes in land cover and land use have been
extensive in the U.S. and in the rest of the world, and these changes have local, regional,
and even global impacts on the hydrological cycle. Among other things, they can
dramatically alter surface properties, affecting the surface heat budget and the
partitioning of precipitation into surface storage or runoff.
Other factors also affect model performance. One is the inhomogeneous distribution of
water vapor. For example, major transports of water vapor from the oceans to terrestrial
regions occur in narrow streams of moisture that vary geographically and with height,
making them difficult to measure, let alone model. Another is the dependence on
timescale. Soil moisture, vegetation, and snow cover can influence the flux of moisture
and energy into the atmosphere depending on the nature of the atmospheric flow, the
surface radiation budget, and surface properties. To improve the reliability of climate
projections a better understanding is needed of which are the key processes to represent
and how processes that occur at scales smaller than the model grid squares interact. In
particular, while some progress has been made in cloud parameterizations, the
representation of clouds and cloud/ precipitation processes remains the greatest
uncertainty in climate models. Furthermore, since cloud processes are inextricably linked
to other critical water cycle processes, improved representation of clouds will be key to
improved simulations as well as climate projections.
Because the set of observations available to answer questions about the natural variability
and change in the water cycle are generally limited in both time and space, new observing
technologies and creative data fusion and assimilation methods will have to be developed
to combine inhomogeneous data with hugely varying temporal and spatial characteristics
into physically and dynamically consistent data sets. This will be true both for existing
and future data sets.

How have the characteristics of the water cycle changed in recent years and is the
number of extreme hydrologic events (droughts, floods, high intensity rain events)
increasing?
To what extent are changes in the water cycle attributable to natural variability as
opposed to human induced change?

DISCUSSION DRAFT
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20

21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38

Products and Payoffs

How are the rates of regional groundwater recharge, soil moisture availability,
and runoff production affected by changing global precipitation patterns,
vegetation distributions and cryospheric processes?
What are the average regional water fluxes between surface and subsurface
arising from recharge and discharge processes and their seasonal and interannual
variability?
What are the factors that control evaporation and evapotranspiration on local and
regional scales, and how are they affected by climate variability and change?
What are the characteristics of, and processes governing, water vapor distribution
and transport in the lower atmosphere and how do they affect precipitation
patterns on short- and long-term scales?
How do aerosols, their chemical composition, and distribution feed back on cloud
formation and precipitation processes and patterns?
What are the characteristics of upper tropospheric water vapor and clouds and
how are they affected by deep convection?
What is the relative importance of local and remote factors in extreme hydrologic
events such as droughts and floods?
In what ways do aerosols affect the hydrologic cycle, particularly the space-time
distribution of precipitation over land?

Documentation of trends in key variables through data analysis and comparison


with model-simulated trends to assess natural variability versus human-induced
changes (5-10 yr.)
Integrated long-term global and regional data sets of critical water cycle variables
from satellite and in situ observations for monitoring climate trends and early
detection of climate change. (2-5 yr.)
Improved regional water cycle process parameterizations based on process studies
conducted over regional test beds to improve the reliability of climate change
projections. (5-15 yr.)
Long term records of flood/drought frequency and intensity from proxy data such
as tree ring data (5-10 yr.)
10-year data set of assimilated estimates of soil moisture and evapotranspiration
rates (2 to 5 yr.).
High-resolution data sets of precipitation amounts, distribution, and intensity over
a regional testbed to be used to develop improved parameterizations of
precipitation processes (5 yr.)
New methods for measurement and estimation of subsurface fluxes (5-15 yr.).

39
40
41
42

Readiness and Feasibility


Techniques for measuring many of the water variables have improved, but the number of
observations is limited and, in some cases, new sensors are needed. A number of new
10

DISCUSSION DRAFT
1
2
3
4
5
6
7
8
9
10

satellite-based sensors are just now or shortly will be available and will provide data on a
number of key variables. In some cases, such as the current soil moisture measurements,
new retrieval methods have produced more accurate estimates. Even with the potential
for global coverage via satellite and ground based remote-sensing platforms, there
remains a critical need for in situ observations of many GWC variables at higher spatial
resolution and at more frequent intervals. Generally, in situ networks are declining and
deficiencies in these networks will inevitably affect the ability to advance the GWC
agenda. Thus, remotely sensed data must be supplemented with data from appropriate
ground-based systems.

11

DISCUSSION DRAFT

Box 1. The 1993 Mississippi River Floods


In the summer of 1993, the Mississippi River basin experienced anomalously high rainfall,
following a winter and spring in which precipitation was generally above normal. During
June and July, an unusually persistent branch of the jet stream was positioned over the

Dominant weather patterns over the United States for June-July 1993 (top panel) and
flooding near West Alton, Illinois, during July 1993 (bottom panel) (USGS, 1993).
12

DISCUSSION DRAFT
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46

Box 2.1 (continued). The 1993 Mississippi River Floods


upper Mississippi and Missouri River basins. This phenomenon was caused by a lowpressure system over the southwestern United States, combined with a stalled highpressure system over the southeast, which created an anomalous low-level flow of
warm, moist air from the Gulf of Mexico that collided with cool, dry air from Canada
over the central states. The result was two months of much above average precipitation.
The combination of the high rainfall with wet antecedent conditions resulted in mean
monthly discharges of the Mississippi River at its mouth during August and September
that exceeded the largest values for the previous 63 years. At 45 USGS stream-gauging
stations over a wide area of the central United States, peak discharges exceeded the 100year flood. Damages exceeded $20 billion, making this one of the most costly natural
disasters in U.S. history. Although the conditions that led to the 1993 flood have been
quite well documented, what is much less well known is the likelihood of similar largearea flooding in the future. The 1993 flood was especially notable because it occurred
during what is normally the low-flow period. Better understanding of the global water
cycle will help to predict the possible occurrence of rare events like the 1993 flood, and
thus to mitigate future flood damages.

Data assimilation is an advanced method for using measurements and models in


combination to provide internally consistent data for analysis. For example, recent
research has produced more effective methods of assimilating remotely sensed data and
of predictions of some GWC variables such as evaporation at regional scales. Newly
developed procedures for assimilating precipitation data for use in land-surface models
are being used to produce experimental high-resolution soil moisture and other land
surface data products on a routine basis. Further advances in data assimilation for other
variables are needed because future large-scale or global observational networks will
consist primarily of remotely sensed data that are augmented with limited in situ
measurements, and there is much promising work going on in this area.
Many of the processes involved in the water cycle occur on scales smaller than are
currently measured on a routine basis or are represented in numerical models. For
instance, cloud formation is a small-scale process that cannot be explicitly represented in
a climate model. Moreover, a full understanding of cloud formation, and all the
processes involved, remains elusive. Model development can be accelerated by
interdisciplinary field studies over regional testbeds that provide much needed
understanding of scaling effects. New parameterizations of water cycle/ climate
feedbacks (e.g., cloud-aerosol and land-atmosphere) and sub-grid scale processes (e.g.,
clouds, precipitation, evaporation, etc.) can be developed and validated on a regional
scale. The sensitivity of global models to these new parameterizations can then be
evaluated. In addition, cloud resolving models are proving to be a useful tool in
ascertaining which processes are important in cloud and precipitation processes.

13

DISCUSSION DRAFT
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45

Research Needs
New observing capabilities, both satellite and in situ, will be critical to detecting patterns
and quantifying fluxes, especially instruments for global measurement of water vapor,
precipitation, and terrestrial water cycle variables such as soil moisture. The decline of in
situ networks needs to be reversed and data sets developed to ensure consistency between
historical and new observations. Network enhancements are needed to obtain data on
critical quantities such as river discharge, precipitation and snow pack in mountain
regions, as well as estimates of the fluxes between the surface and subsurface and
recharge rates at the basin scale. There is a need for new data assimilation techniques
that combine different kinds of data and data with varying spatial and temporal
characteristics to produce consistent data products for research and process studies of key
water cycle variables, such as clouds, precipitation and soil moisture. New models are
needed that can simulate the critical water cycle processes at resolutions that will allow
comparison with long-term data sets. Critical processes include precipitation, water
vapor fluxes and transport at a variety of scale, coupled atmosphere surface (both land
and ocean) interactions, runoff, subsurface water, etc. Finally, process studies to
investigate cloud and radiation processes at small scales will allow development of subgrid parameterizations for climate models.

Linkages
National
As with all parts of the climate system, the factors affecting the water cycle variability in
any location will be a complex combination of local and remote forcing mechanisms
operating on a variety of timescales. Within the CCSP, these water cycle studies will
need to be coordinated with those under the Climate Variability and Change element.
Improvement in parameterizations of water cycle processes will provide input for the
Climate Models and Simulation and Applied Modeling elements.
Other CCSP component programs that will contribute to and/or benefit from these
research efforts include Atmospheric Composition; Ecosystems; Land Use/Land Cover
Change; and Grand Challenges in Modeling, Observations and Information Systems.
International
Key programs with which linkages are being forged include the World Climate Research
Programme (WCRP) (including GEWEX and CLIVAR); the International GeosphereBiosphere Programme (IGBP); various programs of the United Nations (WMO, FAO and
others). In particular, on-going collaborations will lead to development of an IGOSPartners Water cycle theme report to guide the evolution of integrated global water cycle
observing systems.

2.2 WATER CYCLE FEEDBACK EFFECTS ON THE CLIMATE


QUESTION 2: HOW DO FEEDBACK PROCESSES CONTROL THE
INTERACTIONS BETWEEN THE GLOBAL WATER CYCLE AND
14

DISCUSSION DRAFT
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45

OTHER PARTS OF THE CLIMATE SYSTEM (E.G. CARBON CYCLE,


ENERGY), AND HOW ARE THESE FEEDBACKS CHANGING OVER
TIME?
State of Knowledge
Feedback processes are interactions between components of a system as it responds to
inputs. When the global water cycle is considered as a component of the Earth/ Climate
system, feedback processes transmit external drivers, such as the increase atmospheric
CO2 through the system. An input provides a response in one component that, in turn,
triggers a response in another. Mutual adjustments continue to occur and reverberate
through the system. The system may return to equilibrium, develop a new cycle or
cycles, or continue to exhibit chaotic behavior.
Feedbacks can be positive or negative, with positive feedbacks enhancing the initial
response and negative feedbacks inhibiting or counteracting it. In the case of the
feedbacks between the water cycle and the carbon cycle, a positive feedback could
increase the accumulation of carbon dioxide in the atmosphere while a negative feedback
with slow the growth of atmospheric CO2. For interactions with the climate system, a
positive feedback would enhance the global warming effect while a negative feedback
would either slow the effect or produce a cooling. Understanding feedbacks between the
water cycle and other component of the climate system is critical for climate modeling.
There is a great deal of fundmental research to be done in this area.
Depending on time-scale, elements of the global/regional water cycle can act as either
forcings or feedbacks on/within the Earth/Climate system. While all feedbacks are
mechanisms not all water cycle mechanisms (discussed in Question 1) are feedbacks. For
example, the release of latent heat during the formation of precipitation may be
considered as a mechanism or internal process as opposed to a feedback. However, an
initial change in land cover (vegetation, snow cover) due to a forcing, which then causes
an additional change in land cover via the water cycle would be a feedback effect.
One major feedback involves the linkages between the global water cycle and the
greenhouse gas warming. A small incremental increase in temperature will result in
greater rates of evaporation, as a warmer atmosphere is capable of holding larger amounts
of water vapor. The increases in water vapor, a very effective greenhouse gas, lead to a
further increase in temperature. Water vapor, the most important greenhouse gas in terms
of energy absorption (as measured in Watts/m2) and long-wave heating of the planetary
surface, is a major contributor to the net warming effect following an increase caused by
anthropogenically emitted greenhouse gases such as CO2 and CH4. However, there are
serious uncertainties in the vertical and spatial distribution of water vapor that might
result from an initial CO2 (plus other greenhouse gases) warming, and consequently the
net resultant effect of the water vapor feedback. This is a positive feedback that could
continue unchecked if there were no counterbalancing effects in the atmosphere.
However, with increasing water vapor content from increased temperature, the potential
for cloud production increases. Clouds have several impacts. Most significantly, during

15

DISCUSSION DRAFT
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46

the day, they reflect incoming short-wave radiation back to space, thereby reducing the
amount of radiation reaching the surface to warm it. Under these conditions, clouds have
a negative feedback effect. However, under other conditions, the cloud may be trapping
outgoing radiation, leading to a positive cloud feedback. Over the globe, depending on
cloud types, the net feedback effect may be positive or negative. There is evidence to
suggest that the treatment of clouds in climate models is one of the major determinants of
their temperature sensitivity to greenhouse gases. These feedback effects are complicated
by the presence of aerosols in the cloud that act to change the albedo of the cloud and
reduce the likelihood that the clouds moisture will rain out. At present, it is generally
believed that there is insufficient observational evidence to determine whether clouds
have a positive or negative net feedback effect on the climate.
Another important feedback comes from the linkage between the water cycle and the
carbon cycle. Observational evidence indicates that transpiration rates for plants are high
at the same time that carbon dioxide fixing by the plants and hence carbon dioxide flux
from the atmosphere to the plant canopy is large. This connection between transpiration
and the carbon flux suggests that many of the same processes must be controlling the
rates of transfer. It also suggests that common approaches to measuring and
parameterizing water vapor and carbon fluxes may exist. There are other feedbacks
between the carbon cycle and the water cycle. For example, when an environment is
humid, plants will grow more rapidly and draw carbon dioxide from the atmosphere and
release more water to the atmosphere. The addition atmospheric moisture can enable the
production of more clouds and rain, which will continue to moisten the ground and
prolong the growth of plants and the continued transpiration of moisture to the
atmosphere. This positive feedback is one that has not been examined in detail, but may
lead investigators to identify of areas where terrestrial carbon sequestration would be
particularly effective.
Another important feedback effect between greenhouse gas cycles and the water cycle is
expected to unfold when increases in temperature and land use and vegetation cover
changes affect the hydrology of sensitive regions. For example, melting permafrost could
lead to larger areas of standing water at higher latitudes that could increase the rate of
methane production. Methane, a very active greenhouse gas, could lead to further
warming if present in sufficient quantities.
In current global and regional coupled models, feedback processes are typically poorly
represented or accounted for. This difficulty arises from a lack of observations to provide
for a basis for parameterizing these processes and partly due to the complexities involved
in modeling them. In addition, there is a need conduct studies that will clarify the
significance of the various feedback processes and to clarify how the ones that have the
most significant effects for the CCSP operate.
There are also major uncertainties in the cloud/precipitation response to forcings on the
Earth system and in cloud radiation feedback effects. The response of the atmosphere to
increased evaporation at the ocean surface is expected to be an increase in cloudiness.
The effect of clouds on the radiation balance depends on whether the clouds form in the

16

DISCUSSION DRAFT
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44

upper or lower troposphere and whether they form near the equator or the poles.
Furthermore, it is not clear how these clouds will be distributed over the planet nor is it
clear how the production of precipitation from these clouds will be altered as a result of
forcing. The vertical distribution of precipitation formation can have important effects on
the atmospheric heating profiles and on patterns of storm development.
Water has an important influence on atmospheric circulation. Water cools its
surroundings as liquid and solid water are converted into water vapor. Without this
cooling the land surface would warm, much like hot pavement or the sand of deserts. On
the average, this latent cooling is balanced by the latent heat released when water vapor is
converted to liquid and solid cloud particles. Because of the large latent heat involved in
the condensation of water molecules, water vapor is a very effective means of storing
energy and the latent heat flux in the atmosphere is a major component of the overall
transport of energy from equator to poles. In general, latent heat is the principal source of
energy that drives cyclogenesis and sustains weather systems like convective cells that
generate tornadoes and tropical storms that evolve into hurricanes.
Water molecules are strong absorbers of infrared radiation and the resulting greenhouse
effect of atmospheric water vapor is, by far, the strongest determinant of the Earth's
surface climate. Furthermore, atmospheric humidity is highly variable and responds to
changes in atmospheric temperature, thus providing the most effective feedback
mechanism tending to amplify global climate changes induced by other factors.
Furthermore, clouds contribute about 50% of the planetary albedo, and absorption of
terrestrial radiation by clouds is equivalent to that of all "greenhouse gases" other than
water vapor. Radiative heating or cooling is a major contribution to the diabatic processes
that cause air parcels to rise or sink in the atmosphere and, in general, power weather
systems. The net radiant energy that reaches the Earth surface is the source that controls
temperature, drives evaporation, and feeds photosynthesis and the Earth's primary
biological productivity. Being able to measure and to forecast the evolution of the spatial
and temporal patterns in water vapor and clouds is a key to applications of science to
climate, water resources, and ecosystem problems.

Draft Research Questions

What is the sign and magnitude of the cloud-radiation-climate feedback effect and
how does it vary with latitude and season?
How is the water vapor-climate feedback signal changing, and how can these
feedback processes be better represented in global models?
How do changes in water vapor and water vapor gradients, from the stratosphere
to the surface, affect climate variables, such as radiation fluxes, surface radiation
budgets, cloud formation and distribution, and precipitation patterns, globally and
regionally?
What are the variations and changes in freshwater fluxes to the ocean that could
affect the ocean thermohaline circulation and feed back on global/regional
climate?

17

DISCUSSION DRAFT
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43

How do changes in the global/ regional water cycle feed back on vegetative
growth and carbon sequestration?
How do changes in water cycle processes in cold regions feed back on climate
change? In particular, how would warmer temperatures in the Arctic affect
regional hydrology and methane production over northern land areas?

Products and Payoffs

New parameterizations for water vapor, clouds, and precipitation processes for
use in climate models, using new cloud resolving models created in part as a
result of field process studies (2-5 yr; Next generation improvements: 5-15 yr).
Integrated water cycle time series data sets (derived from satellite and surfacebased remote sensing, combined with in situ measurements) of tropical and extratropical precipitation, clouds and cloud properties, aerosols, short wave and long
wave radiation, and water vapor (Initial: 2-5 yr; Periodic improvements and
updates: 5-15 yr)
Enhanced, integrated data sets (remote sensing and in situ) for correlated and colocated studies of the feedbacks and interactions between changes in water cycle
parameters and biogeochemical cycles. Examples include carbon sequestration,
ecosystem impact, and land-use change feedbacks. (Initial: 2-5 yr; Improvements
and updates: 5-15 yr)
Enhanced data sets for feedback studies including water cycle variables, aerosols,
vegetation and other related feedback variables generated from a combination of
satellite and ground-based data to evaluate the role of human influences in climate
change. (5-15 yr).
New models capable of simulating the feedbacks between the water cycle and the
climate system (including biogeochemical cycles) to improve predictions of
climate change and support the development of carbon management strategies (515 yr).
Analyses to identify the variability of cloud and radiation fields and atmospheric
conditions that could be important for cloud feedback research (2-5 yr)Methods to
estimate cloud condensation nuclei (CCN) concentrations.
(2-5 yr)
Studies of the impact of aerosols on cloud drop distributions and the assessment
of the indirect effect. (5 yr)

Readiness and feasibility


Considerable research has been conducted on improving observations of key parameters
of the intertwined global and regional water and energy cycles, as well as research into
improving skill in predicting changes in the variability of water resources and water
availability, including precipitation, evaporation and soil moisture, on time scales up to
seasonal and annual as an integral part of the climate system. Efforts include those of
NASAs EOS program, GWEC, NASAs NSIPP (seasonal-to-interannual prediction) and
DAO (data assimilation and modeling) programs (among others), DOEs ARM program

18

DISCUSSION DRAFT
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33

(observations, process studies, and modeling), GEWEX/GAPP and GAPP-LDAS


programs (NOAA in collaboration with NASA), a broad range of research sponsored by
the NSF, and the efforts of other agencies involved in research and the operational
monitoring of basic land surface and hydrological parameters (USGS, USDA, USFS,
USBR, others).

34
35
36
37
38
39
40
41
42
43
44
45

Research Needs

Activities are ongoing to quantify and improve the understanding and modeling of key
elements and processes of the global and regional water/energy cycles. Measurement
techniques have been developed and/or improved. For example, more accurate
measurements of water vapor were developed through joint NASA/DOE (ARM)
campaigns. Global measurements are available from NASAs satellites. These include
the EOS Terra and Aqua, TRMM (precipitation radar), QuickSCAT (measuring surface
ocean windsimportant for estimates of ocean evaporation, and GRACE (global surface
and sub-surface water availability/resources). Important next generation follow-on
missions include: GPM (Global Precipitation Measurements), Soil Moisture Mission, and
others. Also important are the planned transition of EOS measurements to the NPOESS
system of operational satellites.
The CCSP also supports a number of field testbeds and research basins. They include the
USDA facilities, USGS stream gauge networks, and the DOE ARM sites. Collaborative
efforts are being planned to combine the capabilities of agency resources to provide an
enhanced research capability at existing sites.
Many of the uncertainties in the projections of the warming effects of increasing
atmospheric carbon dioxide arise from the inability to adequately represent the cloud and
water vapor radiative feedback processes in models. Advances have been made in
measuring the Earths surface energy and describing the interactions of water and energy
(heat) in the water cycle. Studies show that in the tropics a decrease in cloud cover
accompanies a warming trend in the region. Studies have provided insight into the
relationship between the physical properties of cirrus ice crystals and meteorological
factors, such as temperature and water amount, and the ability of cirrus clouds to reflect
and absorb energy

Many of the uncertainties in the impacts of changes in climate variability and long-term
global change that have been identified in IPCC reports arise from our inadequate
understanding and inability to adequately model GWC processes as they feed back on the
climate system. The current inability to adequately represent these complex multi-scale
processes in climate models is a major source of uncertainty in long-term climate change
projections, seasonal-to-interannual climate forecasts and their impacts. Model
improvements will be accelerated by interdisciplinary field studies over regional testbeds
that provide much needed understanding of scaling effects. New parameterizations of
water cycle/ climate feedbacks (e.g., cloud-aerosol and land-atmosphere) and sub-grid
scale processes will have to be developed and validated to improve the accuracy of
precipitation predictions and projections generated by climate models.

19

DISCUSSION DRAFT
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39

Examples of key research needs:


Development and implementation of satellite based global measurement systems for
precipitation.
New validated parameterizations of feedback processes that affect precipitation such
as cloud-aerosol, land-atmosphere interactions, and cloud-radiation-climate change.
Sensitivity tests of global models to improved parameterizations of feedback and subgrid scale processes (e.g., clouds, precipitation, land surface processes, etc.)
Scaling results from interdisciplinary field studies over regional testbeds for use in the
development of models.
Development of new observing system capability on a research/experimental basis to
measure global/regional water cycle feedback parameters, including in situ and spacebased and surface-based remote sensing instruments and platforms.
Satellite-based global scale measurements of other key GWC variables including
water vapor profiles, soil moisture/ wetness, and sediment transport. Transition of
proven research/experimental observing instruments/platforms to operational systems
(both remote sensing and in situ observing systems). This includes ensuring that the
measurement of key variables is maintained through transitions in observing systems
(globally and locally).
Development of an integrated Earth system modeling infrastructure relevant for the
testing, validation and use reducing uncertainties in climate change (and variability)
predictions and projections.

Linkages
National
To address GWC feedbacks it will be necessary for this element to work closely with
other CCSP Programs including Climate Variability and Change, Climate Models and
Simulation (including Applied Climate Modeling), Atmospheric Composition, Carbon
Cycle, Ecosystems, Land Use/Land Cover Change, Grand Challenges in Modeling, and
Observations and Information Systems.
International
Key linkages for the GWC program include: the World Climate Research Programme
(WCRP) (particularly GEWEX, CLIVAR, SPARC and CLiC) and the International
Geosphere-Biosphere Programme (IGBP) (particularly the emerging iLEAPS project).

2.3 PREDICTING WATER CYCLE VARIABILITY AND CHANGE

20

DISCUSSION DRAFT
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45

QUESTION 3: WHAT ARE THE KEY UNCERTAINTIES IN SEASONAL


TO INTERANNUAL PREDICTIONS AND LONG-TERM PROJECTIONS
OF WATER CYCLE VARIABLES AND WHAT IMPROVEMENTS ARE
NEEDED IN GLOBAL AND REGIONAL MODELS TO REDUCE THESE
UNCERTAINTIES?
State of Knowledge
Seasonal-to-interannual variability in the global water cycle is largely determined by
ocean and land processes and their impacts on the atmosphere. The prediction of this
variability relies on the persistence or memory in surface conditions that tends to provide
the atmosphere with consistent anomalies in fluxes over periods of weeks or months or
even years. Again, the time scale of memory in the atmosphere is fairly short, but due
to the atmosphere's connection to the land and ocean, each of which is characterized by a
much longer memory. As a result current dynamic global and regional models
demonstrate limited skill in predicting precipitation and GWC variables that strongly
depend on precipitation, such as soil moisture and runoff, on time scales beyond a few
days. Droughts in particular and pluvial periods to a lesser extent can be extended and
maintained at seasonal-to-interannual time scales, with potentially severe consequences
for agriculture and water resources. The El Nio / La Nia cycle is the most obvious
example of a coupled phenomenon that produces significant seasonal-to-interannual
variability. Over land, it is known that soil moisture, groundwater, snow processes, and
vegetation can also contribute to the memory effect. While the large-scale influences on
the atmosphere by major anomalous oceanic variations such as El Nio events have been
well documented, memory effects of land conditions and their consequences for
evapotranspiration and albedo are not fully quantified. In addition, climate models exhibit
serious bias in precipitation due to their inability to fully represent small-scale cloud and
precipitation processes. However, much work remains to determine which variables can
be predicted and which ones cannot. Furthermore, in the context of climate change
issues, the relevant scales are larger and the issues more complex, especially considering
the additional effects of increasing populations, rising standards of living and competition
amongst the users of water.
Due to the large uncertainties associated with the outputs of climate models representing
conditions up to 100 years in the future and the practice of using them as scenarios, the
term projections is used to distinguish such outputs from predictions. One of the
most critical deficiencies in climate change projections involves precipitation and soil
moistureessential parameters for assessments of the nations future water availability.
Advances in land surface models have led to a much improved capability of simulating
coupled atmosphere-land system. When forced by observed precipitation and radiation
data, todays land models are capable of simulating realistic land surface conditions on
time scales much longer than a few days. Using this approach land data assimilation
systems now can produce both real time and retrospective analyses of land surface
variables that can be used for diagnostics, initialization, and validation of coupled
atmosphere-land models for climate prediction and projections. Some improvements have
21

DISCUSSION DRAFT
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45

been realized through the development of precipitation assimilation capabilities. These


successes suggest that more realistic specification of land surface conditions can lead to
better precipitation predictions and increase confidence that improvements in the land
components of models will contribute to better water cycle predictions at daily, weekly
and seasonal time scales. Studies based on both observations and model simulations also
have shown potential linkage of land surface conditions, primarily soil moisture, on
seasonal climate anomalies over the US, such as the major drought in 1988 and the
Mississippi floods of 1993.
Many uncertainties in seasonal to interannual climate predictions and climate change
projections will only be reduced when better representations of GWC processes can be
incorporated into climate models. At seasonal time scales, predictions rely on accurate
specification of the initial conditions and characterization of the boundary layer
conditions that control surface-atmosphere interactions. In particular land surface and
ocean feedbacks to the atmosphere, cloud and precipitation processes, and hydrologic
surface processes need to be addressed. Projections at decadal to centennial time scales
require representations of the boundary that can evolve over time. Surface boundary
forcing changes are particularly important in sensitive areas such as the cryosphere
(permafrost, snow cover and ice cover) and for vegetation conditions and water cycle
variables such as soil moisture.
A critical prediction problem involves advance warning for major flood and drought
events. The development of a capability to reliably assess whether hydrologic extremes
will increase as greenhouse gas concentrations increase is also important. The increasing
property damages from floods suggests that two factors may be at work, namely the
tendency for more people to locate in flood plains and, possibly, a trend towards the
intensification of the hydrological cycle. In order for an extreme event to occur the
following factors are usually be present:
1) large-scale circulation patterns that enhance vertical atmospheric uplift for
floods or increase the stability of the atmosphere for droughts;
2) regional patterns and feedbacks that accentuate the larger scale factors
contributing to floods and droughts;
3) preconditioning of the system to increase the impacts of the flood or the
drought event. For example, antecedent wet soils will leads to enhanced
floods for a given rainfall, while antecedent low water tables and
desiccated vegetation prior to the drought will cause the drought impacts
to be much greater.
The prediction of large floods and droughts requires attention to each of these factors.
Through global and regional climate models, it is possible to have good predictive
skill for some areas, because the atmospheric flow patterns that are frequently
associated with heavy rains or drought events can be identified. However, the
modeling of regional feedbacks requires a good understanding of land atmosphere
interactions, while the modeling of antecedent conditions requires hydrologic and
biospheric models and monitoring programs that will account for the effects of
prolonged rainfall, or lack thereof, in a given region. Regional feedbacks and the role

22

DISCUSSION DRAFT
1
2
3

of antecedent conditions are two aspects of extreme events that are poorly
understood.
Box 2. The Importance of Predicting the Effects of El Nio on North America: the
1997-98 ENSO Event
Among the largest El Nio events of this century, the winter of 1997-98 saw nearly
unprecedented rainfall in several parts of the southwestern and southeastern United States,
rainfall attributed directly to the effects of extremely warm sea surface temperature (SST) in the
tropical Pacific. Unlike other such events, however, the 1997-98 event was relatively well
predicted, both the SST anomaly in the Pacific and its remote effects, especially in the United
States. The figure shows the precipitation prediction for January through March 1998 made by
the U.S. Climate Prediction Center (top) three months in advance of the winter season as well as
the observed precipitation (middle) and the historical expectation based solely on the presence of
El Nio conditions in the tropical Pacific (bottom). As the figure shows, the CPC forecast was
based on the expectation that El Nino would have a major effect on winter precipitation, and their
predictions were quite accurate for many regions of the country. Individuals and organizations in
climate-sensitive locations across the country made use of the forecast information, taking steps
to mitigate the potential costs of El Nino, and thereby substantially reducing El Nios actual
costs.

23

DISCUSSION DRAFT

1
2
3
4
5
6

24

DISCUSSION DRAFT
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43

Draft Research Questions

How predictable are water cycle variables at different temporal and spatial scales?
For different model resolutions, how can key water cycle processes be better
simulated in current climate models in order to enhance their capabilities to produce
more accurate seasonal to interannual predictions of water cycle variables?
How can the representation of water cycle processes in climate change models be
improved to reduce uncertainties in climate change projections of hydrologic
variables? Variations of water cycle on longer time scales are associated with
changes in slowly varying components of the Earth system, such as deep oceans,
glaciers, ice sheets and sea ice, land cover and land use, and atmospheric
composition. The challenge for prediction of these variations depends on our ability
to understand and model the fundamental processes that affect climate change.
How can GWC subgrid scale processes best be characterized in climate models being
integrated over long time intervals?
To what extent will the seasonality, intensity and variability of high latitude
freshwater fluxes (evapotranspiration, runoff) and stores (soil moisture, permafrost)
change as a result of climate warming? How well do climate and hydrologic models
simulate these changes? How sensitive are climate change projections to errors in
represented the processes causing these changes in climate models?
What are the critical hydrological and atmospheric factors that are present in major
flood and drought events that can be isolated, quantified and incorporated into water
cycle prediction methodologies?
How well do current global climate models simulate individual components of the
global water cycle and what are the consequences of the models weaknesses for
current climate projections?
What is the optimum structure for ensemble forecasts (in terms of members, models,
start times) that produce the best seasonal precipitation, streamflow and soil moisture
forecasts?
How can the uncertainty in the prediction of water cycle variables be characterized
and communicated to water resource managers?

Products and Payoffs

New drought monitoring and early warning tools based on improved measurements
of precipitation, soil moisture and runoff and data assimilation techniques to use in
the implementation of drought mitigation plans. (2-5 yr).
A regional reanalysis providing a wide range of daily analysis products at 32 km
resolution for a 25 year period for use in analyzing many features that are absent in
global climate data assimilation products. (2-5yr).
Metrics for quantifying the uncertainty in predictions of water cycle variables and
progress in improving their accuracy and for making forecasts more useful in water
resources management. (2-5 yr)
Downscaling techniques, such as improved regional climate models, that bridge the
disparate spatial and temporal scales between global model outputs and atmospheric,

25

DISCUSSION DRAFT
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42

land surface and river basin processes for improved evaluation of potential water
resource impacts arising from climate change. (5-15 yr)
Field and modeling experiments to study the role of mountain environments on
precipitation and runoff production (2-5 yr)
Improved global and regional climate models with improved representations of the
key processes in the models (5-15 yr)
Improved data assimilation which is benefited from the synergy of improved model
and observations and improved assimilation techniques (5-15 yr)
A long-range prediction capability of drought and flood risk (seasonal to internannual
time scales). (5-15 yr)

Readiness and feasibility


Some effort is being directed currently at the seasonal prediction of the GWCs
variability, mainly in conjunction with climate modeling and numerical weather
prediction centers. Improvements made over the past decade to models, have advanced
the ability to close regional water budgets. For example, when combined with new data
assimilation capabilities the annual water budget for the Mississippi basin can now be
closed to within 15 - 20%. While this progress is encouraging for climate applications, it
indicates that more work is needed on predicting critical variables such as
evapotranspiration before predictions will be adequate for the needs of water resource
managers. The development of a Land Data Assimilation System (LDAS) at both the
regional and global scales has allowed a significant reduction in the errors arising in
initial fields due to the way traditional coupled land-atmosphere 4-D data assimilation
systems (4DDA) in soil moisture, soil temperature, and surface energy fluxes. Another
key to these developments are continued improvements in our understanding of land
surface processes, particularly soil moisture, snow cover and frozen ground effects and
their contributions to the memory effects that are evident in droughts and anomalously
wet periods. However, there are reasons to believe that a modest increase in investment in
this area could accelerate the development of a seasonal prediction capability.
Progress on long-term projections (decadal to centennial) of GWC variability have not
matured as quickly, partly because the long model time integrations required to make
such projections do not allow for the complexities of land surface processes to be fully
incorporated into these models. However, recent advances in the national climate model
development strategy are expected to provide a more efficient structure for improvements
on these time scales. It is anticipated that, with a modest shift in emphasis by both the
climate modeling and GWC communities, it would be possible to make significant
progress in improving the representation of key water cycle processes in global climate
models. However this will need to be a focused effort giving priority to those
hydrometeorological processes (including clouds) to which climate models are most
sensitive.

26

DISCUSSION DRAFT
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32

Research Needs

33
34
35
36
37
38
39
40
41
42
43
44
45

Linkages

The major requirements for improved water cycle prediction capability lie in three areas,
namely: 1) improved specification of initial conditions (including boundary conditions),
2) improved parameterization of relevant physical and biological processes and the land
surface condition, and 3) improved model structure. The third area is addressed in the
Climate Modeling section. Addressing land surface-atmosphere interactions will require
research on the entire coupled system including cloud and precipitation feedbacks, the
interactions of the lower boundary layer with land and ocean surfaces, and the role of
groundwater- surface water and biospheric interactions. In addition, data sets are needed
for evaluating and testing the water cycle components of coupled models, especially soil
moisture and regional evaporation and for the improvement of regional downscaling and
statistical forecasting techniques. Advances in prediction capabilities will depend on
improvements in model structure and initialization, data assimilation, and representations
of the key water cycle processes in models.
In summary the following developments are needed:
Observations: Improved observations (both ground-based and satellite) of water
cycle variables and fluxes (such as temperature, precipitation, snowpack, soil
moisture, vegetation properties, radiation, wind, evaporative flux and humidity) will
provide the foundation for improved predictions of water cycle variables. Enhanced
data sets are needed to evaluate models, to characterize and reduce uncertainties of
model predictions, to improve model initializations and to improve process
understanding.
Predictability studies: Predictability studies will be required to determine the regions,
seasons, lead times and processes most likely to provide additional predictive skill,
and to guide the development of models on all scales.
Process studies and model improvements: Better understanding through field
experiments and modeling studies of less-well-understood processes, such as the
seasonal and longer term interactions of mountains, oceans, the cryosphere and soilvegetation with the atmosphere are needed. In addition, these processes must be
more realistically represented in models at appropriate scales.

National
Seasonal water cycle variability over the US is influenced by both the land and ocean
surface conditions. The relative roles of these influences depend on location and season.
Consequently, the GWC program in collaboration with the Climate Variability and
Change element must address both ocean-atmosphere coupling and land-atmosphere
coupling in order to develop reliable water cycle predictions on seasonal and longer time
scales. In addition, the hydrologic aspects of these predictions require a focused effort
that will involve joint studies with the land use change and ecosystems groups.
International
The GWC research on prediction and predictability also needs to maintain close linkages
with the International GEWEX program and CLIVAR under WCRP and the land
27

DISCUSSION DRAFT
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16

components of the IGBP, as its new research agenda is consolidated. Some of the
hydrological data and modeling issues may also be developed in collaboration with
UNESCOs hydrology program.

17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44

State of Knowledge

2.4 THE CYCLING OF WATER AND OTHER BIOGEOCHEMICAL


CONSTITUENTS
QUESTION 4: HOW DO THE WATER CYCLE AND ITS VARIABILITY
AFFECT THE QUALITY OF AVAILABLE WATER FOR HUMAN
CONSUMPTION, ECONOMIC ACTIVITY, AGRICULTURE, AND
NATURAL ECOSYSTEMS; AND HOW DO THE VARIABILITY AND
INTERACTIONS WITHIN THE WATER CYCLE AND BETWEEN THE
OTHER BIOGEOCHEMICAL CYCLES AFFECT SEDIMENT AND
NUTRIENT TRANSPORT, MOVEMENTS OF TOXIC CHEMICALS, AND
OTHER BIOGEOCHEMICAL SUBSTANCES?
The essential role of water in sustaining all forms of life (plant, animal, and human) and
the enormous contribution of water to economic development throughout human history
have been recognized for millennia. During the past century, major advances have been
made in quantifying the cycling of water between the atmosphere, land areas, oceans,
lakes and streams, and groundwater aquifers. However, there is a consensus in the
scientific community that the current level of understanding of the fate and movement of
water and sediment in watersheds remains inadequate. The many societal problems that
we face related to water availability, use, control, and management are not new. In the
context of climate change issues, though, the relevant scales are larger and the issues
more complex considering the additional stresses of increasing populations, rising
standards of living, and the many diverse and competing demands for fresh water.
As water cycles through the environment, it interacts strongly with other biogeochemical
cycles, notably the carbon, nitrogen, and other nutrients. Flowing water also erodes,
transports, and deposits sediment in rivers, lakes and ocean, affecting the quality of the
water. Soil erosion may result in degradation of farmland through loss of organic soil
material, soil salinization, and gully formation, loss of aquatic habitat, and lost water
storage capacity due to the sedimentation of reservoirs. Soil erosion and sedimentation of
our rivers, deltas and channels also may lead to increased flooding and costs associated
with dredging and maintaining shipping lanes. Toxic chemicals, pesticides, and
agricultural fertilizers in water are detrimental to the health of inland aquatic ecosystems
and coastal zones. The transport by water of bacterial contaminants from sewage and
agricultural waste systems, often under the stress of flooding conditions, exacerbates
human health problems and stresses water management and treatment systems.

28

DISCUSSION DRAFT
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52

Box 3. Recent Notable Oxygen Losses from Important Coastal Waters

Oxygen depletion results from the combination of several physical and biological processes. In Gulf of
Mexico waters (graphed above), hypoxia results from the stratification of marine waters owing to
Mississippi River system freshwater inflow and the decomposition of organic matter stimulated by
Mississippi River nutrients. As a general rule, nutrients delivered to estuarine and coastal systems support
biological productivity. Excessive levels of nutrients, however, can cause intense biological productivity
that depletes oxygen. The remains of algal blooms and zooplankton fecal pellets sink to the lower water
column and seabed. The resulting depletion of oxygen during decomposition of the fluxed organic matter
exceeds the rate of production and resupply from the surface waters, especially when waters are stratified.
Stratification in the northern Gulf of Mexico is most influenced by salinity differences year-round, but is
accentuated in the summer due to solar warming of surface waters and calming winds. Oxygen depletion
follows a fairly predictable annual cycle, beginning in the spring, and becoming most widespread,
persistent, and severe during the summer months.
Midsummer coastal hypoxia in the northern Gulf of Mexico was first recorded in the early 1970s. In recent
years (1993-1999), the extent of bottom-water hypoxia (16,000 to 20,000 km2) has been greater than twice
the surface area of the Chesapeake Bay, rivaling extensive hypoxic/anoxic regions of the Baltic and Black
Seas. Even in 1998, the hypoxic area covered 12,400 km2, an area about the size of Connecticut. Prior to
1993, the hypoxic zone averaged 8,000 to 9,000 km2 (1985-1992).
Source: Nancy N. Rabalais, Louisiana Universities Marine Consortium, 8124 Highway 56, Chauvin,
Louisiana 70344 (http://www.csc.noaa.gov/products/gulfmex/html/rabalais.htm)

29

DISCUSSION DRAFT
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46

Water conservation, water use efficiency, and watershed management play critical roles
in the development of cost-effective solutions to social, economic, and environmental
problems caused by water scarcity, water stress, and extreme climatic events. An
improved understanding of the hydrologic processes that determine the fate and
movement of water in watersheds is critical to the development of effective strategies for
water management in a changing environment.
Reliable techniques are needed to quantify water resource responses to climate forcing
and to use this information to predict the hydrologic consequences of climatic change.
This research would require a major long-term commitment of resources to monitor water
and energy fluxes over a range of scales and to develop accurate, cost effective methods
to characterize watersheds and simulate their integrated responses at the regional and
continental river basin scales. Currently available water cycle predictions and projections
are inadequate for such assessments.
Watersheds are the fundamental landscape units used to determine direct and off site
impacts of agricultural food and fiber production systems, soil and water conservation,
climatic variability and watershed management practices. Watershed responses (e.g.,
hydrologic, riparian, water quality) to management and climatic inputs result from the
complex interactions among numerous watershed factors and attributes (e.g., soils,
vegetation, land use, management practices) and vary across geographic regions. Many
of the major challenges in scientific hydrology and watershed management relate to
quantifying the space-time variability of both weather and climate data, watershed
characteristics, and tracking the ever-changing land use and management practices within
the Nations agricultural and natural watersheds. However, science has not reached the
point where these complex responses can be fully quantified and predicted at the
temporal and spatial scales necessary to provide information with the accuracy desired by
water resource managers and policy makers. Furthermore, the responses of basins to
water cycle variability is also dependant on the basin characteristics such as topography,
land cover, soils, and the types and level of socio-economic activity within the basin.
In order to gain the understanding needed to integrate water resource and water quality
management under conditions of climatic change, researchers need to focus on how
processes within this interactive system affect water availability and water quality and
how these impacts vary over time. Systematic monitoring of flows and the fate and
transport of nutrients, chemicals, and pathogens within our Nations rivers and aquifers is
needed to acquire the basic information necessary to understand and predict the effect of
climatic change on our water resources. Effective use of this information will require
research on the causes for observed patterns. Past researchers have lacked access to
spatially and temporally distributed data and have been forced to combine multiple
pathways and reservoirs of many sizes and shapes in "lumped" models to simulate
watershed responses at regional and continental scales. These models are inadequate for
predicting and simulating the fate and transport of nutrients, chemicals, and pathogens,
because the hydrologic analysis needs to be connected with ecological, chemical,
microbiological processes as well as social and economic processes. A new generation of
distributed models that is able to utilize the full information content of distributed data is

30

DISCUSSION DRAFT
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45

now in its early developmental stages. These models use characterizations of pathway
types to capture chemical and biological processes that can scale from a farm field to a
river basin. The next step is to develop methods for scaling up these processes so their
cumulative impacts can be determined at watershed and larger scales.
At present, a major obstacle to distributed modeling advances is the absence of a
monitoring framework needed to generate the data at sufficiently high spatial resolution
to allow evaluation and testing of spatially distributed land surface models. The data
bases that are needed include critical variables such as surface runoff, aquifer depth,
chemical and biological processes that alter the quality of the flowing water, and factors
leading to changes in water demand (associated with population growth and higher
standards of living) and their hydrological impacts, land use, agricultural practices, and
climate. Data are needed to quantify flows by pathway, to determine chemical and
biological changes along pathways, and to assess how these changes, over time, alter the
physical system as well as ecosystems the flows supports. Observational networks in the
major hydroclimatic and agroecological regions of the country are required to make
simultaneous uniform, long-term, consistent, high quality data sets for tracking the water,
energy, and biogeochemistry cycles. These networks should have an oversight
mechanism that builds upon existing experimental watersheds and provides a unified
framework to understand how climate change will affect critical national resources of
water (quantity and quality), carbon, and nutrients.
The data bases described here will enable the scientific community to address the
following issues:

Effects on water quality of atmospheric deposition. Land use changes alter loadings
of atmospheric vapor, dust, and chemical and other aerosols that are later deposited
on land to become major non-point sources of pollution.
Loadings of non-point source pollution washed by storm runoff into streams and
carried by infiltrating water from urban areas or from fields treated with agricultural
chemicals into aquifers.
Performance of soils, buffer strips, wetlands, detention ponds, and other natural
environments in ameliorating pollutant loadings through diverse weather sequences.
Determining if soil erosion rates exceed soil production rates, and whether current
land management practices are sustainable.
Transport of pollutants and nutrients through rivers, lakes and estuaries.
Effects of deforestation, agricultural practices, fires, urbanization, and other land
changes on sedimentation and on fresh water quality, quantity, and distribution.
Subsurface changes in water chemistry related to climate and other stresses at the
surface. Engineers can use information on change processes to contain and degrade
toxic materials, reduce subsurface contaminant discharge into rivers and estuaries,
and foster use of alluvium and riparian ecosystems in water pollution control.
Effects of groundwater fluxes on terrestrial, riverine, and coastal ecosystems and
ultimately on geochemical balances at all scales.

31

DISCUSSION DRAFT
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22

Draft Research Questions

23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43

Products and Payoffs

How does soil erosion and sediment transport from the farm field to stream to
watershed scale and through entire river systems vary as function of hydrologic
processes and basin characteristics?
How do changes in climate and land cover alter runoff quantities and hence the
transport of sediments, nutrients, and other chemicals?
How do changes in climate, land cover, and nonpoint waste sources alter water
quality in streams and aquifers?
How do physical processes in streams and aquifers change the quality of water
available for human uses and natural ecosystems?
How does water cycle variability and change affect the transport of nutrients in major
rivers and influence the formation of hypoxia zones in the estuary areas?
How do physical, chemical, biophysical, and microbiological processes interact along
upland and stream field pathways to alter water quality? Systems of primary interest
for effects on stream chemistry are stream alluvium (the hyporheic zone) and
hillslope soil (the vadose zone).
How can the linkages between particle and chemical transport be quantified as
chemicals are adsorbed and desorbed in mixing zones with a wide variety of
characteristics?
How do field scale interactions accumulate to change water quality and quantity at
watershed and larger scales?

Protocols for establishing commensurate sets of reliable benchmark data on surface


water, ground water, sediment, toxic substances, and biogeochemical constituents at
watershed and river basin scales for multidisciplinary studies aimed at improved
integrated watershed management. Existing protocols will be reviewed and modified
as necessary, existing study areas will be considered for continuation and
enhancement, and additional study areas will be selected. (2-4 yr)
Nationally consistent assessments of the water-quality conditions in our Nation's most
heavily used streams and aquifers, trends in those conditions, and the primary natural
features and human activities that affect them. (2-4 yr)
Intensive field- and watershed-scale investigations of distribution, transport, fate and
effects on contamination by toxic substances at local releases and non-point sources
in order to provide objective scientific information to improve characterization and
management of contaminated sites, to protect human and environmental health, and
to reduce potential future contamination problems. (2-4 yr)
Long-term monitoring and analysis of stream flow and water quality in areas that
have been minimally affected by human activities, in agricultural watersheds, and in
the Nation's largest rivers in order to characterize time trends and spatial patterns of
regional flow and water quality variability, as well as concentrations and fluxes of
sediments and chemicals, associated with both natural processes and regional-scale
societal impacts. (2-4 yr)

32

DISCUSSION DRAFT
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20

21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45

Readiness and feasibility

Integrated remote sensing and ground-based observations of the water, carbon, energy
and nutrient cycles to develop and validate remotely sensed algorithms to enable
extrapolation to regional and continental scales. (2-4 yr)
Improved distributed models that partition precipitation among evapotranspiration,
surface, and subsurface pathways. Characterize water by its chemical and
microbiological characteristics, route flows, and quantify physical and chemical
interactions for evaluating impacts of climatic change on water quality and nutrient
cycling. Programs for national water quality assessment and studies into the fates and
effects of toxic substances will receive powerful new tools. (5-15 yr)
Improved modeling and remote sensing methods for scaling up from individual
pathways and mixing zones to collective performance in watershed systems. (5-15 yr)
Methods for using ground and remotely sensed observations together with models to
gain better understanding of watersheds as hydrogeochemical units. Progress will
lead to feedbacks to refine and improve the database and the models and to establish
more cost effective methods for sustaining water quality and water availability. (5-15
yr)
Decision support systems and recommendations for sustainable management of water
and land resources that account for changes in the environment, provided economy
opportunities, and meet societal needs. (5-15 yr)

Numerous recent field and modeling studies have investigated the transport and
accumulation of pollutants in streams and ponds, particularly in bed sediments, and
subsequent movement of nutrients from land through coastal wetlands. Based on the
field results, models are being developed that simulate how aeration, binding, pollution,
and storm energy interact determining the contribution or river transport to pollution
loading in coastal waters. Integrated field and modeling studies in hill-slope
environments have elucidated flow paths and chemical reactions in the dynamic mixing
zones near the soil surface. Water quality in streams and rivers is largely determined by
chemical weathering and reactions along runoff flow paths down and through hillslopes
and floodplains. Research producing better definition of these pathways and temporary
storage locations; their physical, chemical and microbiological characteristics, is building
the science needed to predict the impacts of land and its uses and of climate change on
water quality and aquatic ecosystems. A strong base of data and information, including
instrumentation and techniques for observation, sampling, and data dissemination, is
being built by ongoing programs such as the National Water Quality Assessment
(NAWQA), National Stream Quality Accounting Network (NASQAN), Water, Energy,
and Biogeochemical Budgets (WEBB), and Toxics Substances Hydrology programs of
the USGS and by similar programs of USDA and other federal agencies. The data and
knowledge developed by these programs will serve as a sound foundation for
development of research to answer the many remaining questions about the effects of
water cycle variability and potential change on water quality. As the reliability of water
cycle predictions and the resolution of climate change projections increase, the ability to
assess the consequences of climate variability for the cycling of sediments, carbon,
nitrogen and other biogeochemical constituents will increase.
33

DISCUSSION DRAFT
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45

Research Needs
Existing water cycle research largely utilizes estimates of precipitation over watersheds
and stream flows recorded at gages. Simultaneously, basin water quality studies are
largely based on data collected through field sampling over short periods (with the
connections to precipitation and flow data often poorly defined). Measurement
techniques, databases, and procedures for data analysis largely serve a particular
discipline rather than support inter-disciplinary research. As a consequence, we lack
reliable means to quantify associations between flow and water quality, evaluate threats
to natural systems and the sustainability of water supplies, or determine the impacts of
the many forms of global change on water, land, or people.
Better and institutionalized coordination of inter-agency and inter-disciplinary activities
and programs is needed to address this scattered approach. To make advances in our
ability to effectively conduct watershed management assessments it is essential to
maintain a geographically diverse, long-term experimental research and observational
watershed programs.
The first step is adoption of a common vision of the research strategy needed to make
progress in resolving the driving issues. That strategy will include:
Continued maintenance and upgrading of existing hydrologic and geochemical
monitoring networks. Enhanced availability of the resulting data streams to
researchers and managers. Examples include the USGS Hydrologic Benchmark
Network, the USDA-ARS watershed network, and the USGS National Stream
Quality Accounting Network, all of which are being used to gain better information
on time trends and spatial patterns of water quality and quantity variability, and how
natural processes and regional-scale anthropogenic activity drive the variation.
Expansion of monitoring networks into areas identified as particularly sensitive to
climate variations and changes.
Greater use of chemical and isotopic tracers to identify important flow paths (either in
terms of volumes of water transported, contributions to sedimentation, or sources of
pollution plumes). Tracers can also be used to estimate residence times and predict
movements and composition changes in pollution plumes.
A more distributed monitoring approach to catchments hydrology that captures both
pathway flows and chemical and microbial characteristics of solid materials at
boundaries. New instrumentation will need to be developed.
Monitoring that records fluxes, properties of both the flow and the media, and
changes to both systematically and in a framework with common spatial and temporal
referencing.
Hypothesis testing to determine how water flowing along characterized pathways
changes in quality and how consequent feedbacks alter the properties of pathways
themselves.
Methods for scaling up from individual pathways and mixing zones to collective
performance in watershed systems.

34

DISCUSSION DRAFT
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25

Linkages

26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44

State of Knowledge

National
GWC activities addressing this question will liaise closely with other integrated
CCSP/GCRP components, especially Carbon Cycle; Ecosystems; Land Use/Land Cover
Change; Observations and Information Systems programs. US Government agency
programs with which linkages are maintained include the USGS, USDA, EPA, USACE,
Bureau of Reclamation, NSF and others.
International
Critical international linkages for this research include programs under the IGBP
(particularly the emerging land program), UNSECO and WMO (through Hydrology for
Environment, Life and Policy), other programs under the United Nations including
WCRP, FAO and others, and the Global Water Partnership.

2.5 MORE EFFECTIVE WATER MANAGEMENT THROUGH WATER


CYCLE SCIENCE
QUESTION 5: WHAT ARE THE CONSEQUENCES, AT A RANGE OF
TEMPORAL AND SPATIAL SCALES, FOR HUMAN SOCIETIES AND
ECOSYSTEMS OF GLOBAL WATER CYCLE VARIABILITY AND
CHANGE? HOW CAN THE RESULTS OF GLOBAL WATER CYCLE
RESEARCH BE USED TO INFORM POLICY AND WATER RESOURCE
MANAGEMENT DECISION PROCESSES?

Variability and changes in the water cycle lead to profound impacts on human societies
and ecosystems, but many of the linkages between GWC changes and societal outcomes
are not yet understood in the detail needed for formulation of policy and management
responses. The ability to estimate and predict the quantity and timing of streamflow is
essential to planning and operation of water supply, energy generation, irrigation, and
transportation systems. Changes in water availability, water quality, and in some cases,
water temperature, impact domestic, industrial, agricultural and recreational uses of
water, as well as on habitat protection and conservation of ecosystem values. Planners
and managers who must deal with these impacts will require information on the nature of
the potential changes and consequent impacts. Extreme events, such as floods and
droughts, have vividly demonstrated impacts on property, productivity and health. What
will it mean locally, regionally, nationally and globally if the frequency and intensity of
extreme event increases due to an intensification of the water cycle that some studies
suggest? Water management issues frequently arise because of variability in the natural
system or inadequate planning and management decisions by water resource managers or
both. As global water cycle changes are added to the multiple stresses of other global
changes that include a growing population and large-scale land use/cover change,
traditional strategies for managing water supply and related agricultural and natural

35

DISCUSSION DRAFT
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46

ecosystem issues are becoming inadequate, and improvements in prediction are becoming
critical. (Hornberger et al., 2001)
To achieve optimal use of the Earths vital freshwater resources under conditions of
climatic change, scientifically based procedures are needed to assess the economic and
environmental consequences of different water resource management strategies and
policies at the farm, ranch, and regional river basin scales. Many individual impact
studies have been undertaken, and significant progress has been made in some areas. For
example, studies using projections of global warming to evaluate the impacts potential
changes in the behavior of snow packs in the western U.S. have shown that current
infrastructure and operating policies would be inadequate for dealing with the projected
hydrologic regime. However, science is far from achieving the kind of comprehensive
understanding needed to guide resource management and policy decisions.
For instance, a critical need for decision support guidance exists in the area of hydrologic
design and water resources planning. Virtually all design and planning is now based on
the use of statistical frequency analysis, critical period analysis and related methods that
design or plan for the future based on what amounts to extrapolation of historic
observations. Examples include estimation of flood plain extents (typically using 100
year recurrence intervals estimated solely from past observations), flood spillway design
and redesign based on probable maximum flood analysis, and reservoir operating
procedures based on, for instance, historic drought and/or flood occurrences. Long-term
climate variability and potential change calls into question the practice of excluding
information about likely future conditions that may well differ from what have been
observed in the past. Although there have been numerous sensitivity studies
demonstrating the likely impacts of climate change on the performance of water
management systems, there are essentially no methods for including such information in
practice. This disconnect is not limited to questions associated with long-term climate
change. In the realm of seasonal to interannual forecasting, there now exists sufficient
understanding to identify differences in short-term risk associated with, e.g., climate
teleconnection information like ENSO, the Pacific Decadal Oscillation, and the North
Atlantic Oscillation. This information is rarely included in water resources
decisionmaking however, which instead typically utilizes seasonal operating methods that
essentially weight all possible future conditions equally. The reason for these failures to
incorporate scientific advances in planning and management of water resource systems is
sometimes attributed to the uncertainty about future climate. However, water resource
managers routinely incorporate other sources of uncertainty about future demands, and
hydrologic uncertainty within the range observed in the past in their decision-making.
Current water cycle research programs have the potential to break this logjam, but, as
noted in the report by Hornberger et al. (2001), it will require implementation of a new
knowledge transfer framework.
More generally, water management decisions are often constrained by laws, agreements
and societal pressures such as stringent flood control standards, federal and state
environmental regulations, hydropower production schedules, and increasing water
demands for irrigation, urban, industrial and recreation. Recent research results indicate

36

DISCUSSION DRAFT
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47

that water cycle information as well as predictions and analysis tools, can contribute to
the decision-making capacities of water managers who must operate within these
constraints. However, factors such as regulatory inflexibility, institutional structures, and
time pressures make it difficult to change established management and decision systems
to take maximum advantage of new products and tools. In addition, there is a mismatch
between research products and operational information needs. Information is only of
value (1) if decisionmakers understand the implications and uncertainties of the new
information, (2) if they trust the information enough to incorporate it into their plans and
decisions, and (3) if they have mechanisms for responding to the information.
Efforts to eliminate the barriers between research and research users have been initiated
and indicate that early collaborations and side-by-side demonstrations may be effective
tools for speeding innovation. Collaboration is essential to ensure that learning occurs on
both (or all) sides of the research development process. Decisionmakers need to be able
to understand the value of the new information and how it is likely to improve their
decisions. In addition, they must develop an understanding of uncertainty and its
implications. They also need to understand the limits of water cycle science and the kinds
of questions for which it can provide answers. Studies have shown that decisionmakers
are more likely to use water cycle change information if they gain experience with the
use of shorter-term hydro-climate predictions and are provided with mechanisms for
incorporating the information into decision processes. Development of these
mechanisms and experience in their use requires close interactions between the hydroclimate scientists and resource managers. For their part, scientists should learn from these
interactions about the system of constraints under which decisionmakers operate. A
better understanding of the applications environment can lead to modifications in
research design that speed the adoption of results without detracting from the science.
A major deficiency precluding the use of water cycle predictions is that many of the
water cycle related forecasts are either temporally or spatially too coarse and lack
accuracy. It is essential that interactions between decisionmakers and research scientists
identify effective methods for providing this uncertain information in a tractable manner
for decisionmakers. In some cases, more finely detailed climate information or
predictions may be possible; in others, decision-makers may need to work with watercycle researchers to reframe methods and issues in forms that can plausibly be addressed.
The advances needed to overcome the above deficiencies and limitations have been
outlined in the previous four questions.
Another problem limiting our capability to fully assess the adequacy of water resources
for the next ten to fifty years arises from the gaps in socio-economic data such as
historical sequences and patterns of water use and consumption and their responses to
historical ranges of price and water law scenarios, the impact of alternative water law and
water trading strategies on regional development. While the development of models to
assess vulnerabilities arising from changing demands patterns and climate variability
have been developed elsewhere, the US is lagging other developed countries in this area.
Creation and use of models that combine physical data and processes with behavioral and
social factors and processes are hampered by the lack of an integrating framework for
currently incommensurable measures. Furthermore, as Hornberger et al., (2001) noted, a
37

DISCUSSION DRAFT
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40

research effort is needed that integrates advances in physical water cycle science with
social science research to determine how the new information can be of value in policy
makers and operational water managers.

Draft Research Questions

What are the consequences for existing water management infrastructure of


variability and change in key water cycle variables such as evaporation and
streamflow?
How have water consumption patterns and trends changed as a result of major
climatic events, technological innovations and economic conditions? How are
patterns in water consumption likely to change as a result of projected changes in
temperature, land cover/ land use, demographics, water policies, and economics?
To what extent can changes in the management of water resources increase the
adaptability of existing infrastructure to the effects of variability and change in key
water cycle variables?
What are the limits of accuracy for water cycle predictions at spatial and temporal
scales required for water resource management?
To what extent can improvements in seasonal precipitation and streamflow forecasts
improve the management of water reservoirs?
How can water cycle research products, such as the hydro-climatological projections
and forecasts from global and regional climate models, remote sensing data streams,
meteorological and hydrologic monitoring, and snow pack information, be deployed
to improve policy decisions and water resource management?
How can the procedures used to develop design statistics be modified to
accommodate the non-stationarity of the climate?
How can changes in the quality and quantity of water flowing within riparian and
coastal environments arising from land management and policy decisions affect the
provision of environmental services?
What institutional and technical issues limit the use of hydroclimatic predictions by
current water resource agencies? What technical or institutional changes are needed
to encourage improved use of predictions of water cycle variables?
How do variations in water-resource availability over a range of temporal and spatial
scales affect the suitability of existing institutional arrangements, management
practices and the ability to meet existing and planned water allocation commitments?
How can institutions incorporate projections of water cycle change during the next
century in their planning and operations?
How can water cycle, climate information, and predictions be designed and
communicated to be of the most relevance, usefulness, and benefit to decision and
policy makers?

38

DISCUSSION DRAFT
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42

Products and Payoffs

Technology transfer and enhanced capability to produce operational streamflow


forecasts over a range of spatial and temporal scales (days, weeks, months and
seasons), for more effective water management decisions. (2-5 yr)
Assessment reports on the status and trends of water flows, water uses, and storage
changes for use in analyses of water availability. (2-5 yr).
Improvements access and availability of water cycle and climatic monitoring data
series and climate-change projections at the finest temporal and spatial scales
available for use by land and water management agencies, with improved tools for
analysis and use of these products. (2-5 yr)
Report for stakeholders and decisionmakers on general conclusions,
recommendations about information needs and research gaps regarding measures to
mitigate climate change impact (2-4 yr).
Decision support tools integrating historic climate variability, water cycle predictions
and socio-economic analyses to produce planning and management tools that include
these major decision factors. Decision-tree analyses of current decision-making in
selected key water and land management institutions to reveal opportunities and
limitations for the use of water cycle and climatic information. Forecast evaluations
that link levels of error and uncertainty with potential consequences for specific water
use sectors. (2-15 yr)
Integrated models of total water use and consumption for incorporation into decision
support tools that identify water scarce regions and efficient water use strategies. (515 yr).
Development of tools and applications to enable the analysis of impacts of climate
change on water resources and their management. Near term studies include
assessments of climate change on 1) sea level rise impacts on drinking water systems
in Florida; and 2) wastewater treatment costs in the Great Lakes region. (2-4 yr).
Decision support tools for water management decisions such as web-based calculator
for estimating soil retention potential of riparian buffer strips in different locations,
soil types, and plant communities. (2-5 yr).
Watershed and River System Management decision support systems to help resource
managers achieve an equitable balance among competing uses: municipal, fish and
wildlife, agricultural, recreational, hydropower, and water quality. Payoffs include
improved methods and tools for integrating meteorological data (both in situ and
remotely sensed), hydrologic observations, and watershed and river models, for
water-supply management simulations and decision-making for major river basins in
he western U.S. (5-15 yr).
Observing system simulation and forecast demonstrations using advanced watershed
and river system management models and decision support systems, to facilitate
acceptance and utilization of these advanced technologies for improved hydropower
production and river system management. (5-15 yr)

39

DISCUSSION DRAFT
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20

Readiness and Feasibility

21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45

Research Needs

This area of research is ready for development and implementation, particularly in the
area of decision support tool development and application. Seasonal forecasts with
useful skill are available for the western USA to be used in demonstration projects. Some
studies demonstrating the usefulness of these forecasts and how they need to interact with
decision support tools have been carried out and show considerable promise. Results to
date have shown that the methodologies and priorities are strongly regional and require
continued development on a regional basis.
Collaborative exercises were developed whereby water managers and water users test
experimental products developed by the Water Cycle program in parallel with normal
operations to evaluate product utility. In particular, experimental Land Data Assimilation
System (LDAS) products have been developed that will be compared with traditional
algorithms used to determine releases from multipurpose reservoirs in the Upper
Columbia Basin. Other test sites include the Madison and Jefferson headwaters basins of
the Upper Missouri. In addition, a framework for scientist-stakeholder interaction to
improve water management was designed for the headwaters of the Red River in
southwest Oklahoma in conjunction with the Hydrology for Environment, Life and
Policy program.

In order to make accurate assessments of the consequences of GWC variability and


change for water resources, it will be necessary to integrate data from a broad range of
sources and disciplines. Frameworks such as data assimilation and fusion techniques,
Geographical Information System capabilities, and decision support tools are needed to
integrate this information for water resource managers. It will also be necessary to
inventory existing data sources and regional and sector studies, especially for data for
which regional, national, and global repositories are rare or non-existent, such as for
water demand, diversion, use and consumption. A more scientific basis and method for
estimating and predicting water demands is a particularly pressing need.
A critical need for decision support guidance exists in the area of hydrologic design and
water resources planning. Virtually all design and planning is now based on the use of
critical period analysis and related methods. Examples include estimation of flood plain
extents (typically using 100 year recurrence intervals estimated solely from past
observations), flood spillway design and redesign based on probable maximum flood
analysis, and reservoir operating procedures based on, for instance, historic drought
and/or flood occurrences.
Although there have been numerous sensitivity studies demonstrating the likely impacts
of climate variability and change on the performance of water management systems,
there are essentially no methods for including such information in an operational system.
In the realm of seasonal to interannual forecasting, there now exists sufficient
understanding to identify differences in short-term risk associated with teleconnection
patterns such as ENSO and the Pacific Decadal Oscillation. However, this information is
40

DISCUSSION DRAFT
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35

rarely included in water resources decision-making. Current water cycle research


programs have the potential to connect researchers with users, but, as noted in
Hornberger et al. (2001), it will require implementation of a new knowledge transfer
framework. Furthermore, for scientific information to have an impact, it must be
adaptable and timely in the users decision-making process.

Linkages
National
To address the issues that are raised when the basic science research and results begin to
be translated into innovations in water management there will be a need to extend and
enhance the ongoing linkages between the several federal programs that are addressing
and improving the climate-society connections. Linkages between NOAAs Hydrology
research efforts and stream flow forecasting centers, and Regional Integrated Sciences
and Assessments teams (RISA) with NASAs Regional Earth Science Applications
Centers (RESACs), the National Science Foundations (NSF) Science/Tech Centers, the
United States Geological Survey, the Bureau of Reclamation, the Department of Energys
(DOE) Accelerated Climate Prediction Initiative (ACPI) initiative, USDA-Natural
Resources Conservation Service Hydrology research, and the United States Geological
Survey are developing and will be further coordinated through the Water Resources
Research Coordinating Committee. Potential for new collaboration exists between the
Land, Sea, and Space Grant Extension and Research entities under the auspices of the
recently developed Earth Grant initiative.
International
Bilateral collaboration is occurring between the USA and Canada and Mexico through
studies with the International Joint Commission (IJC) related to the effects of climate
change on water resources. In addition, the water cycle program will collaborate with the
WCRP/ IGBP/ IHDP/ Diversitas Joint Water Project, with the WMOs Hydrology and
Water Resources Programme, and with UNESCO through its International Hydrology
Program and the Hydrology for Environment, Life and Policy (HELP), and the Dialogue
on Water and the third World Water Forum. Also, the global water cycle program will
contribute to work through bilateral treaties, particularly with countries like Japan, that
have placed a priority on Water Cycle research.

3. Summary
36
37
38
39
40
41
42
43
44

Important and urgent problems face society as a result of decreasing access to fresh water
arising to possible climate change impacts, increasing water consumption, changing land
use and pollution effects. These problems are the Global Water Cycle are numerous and
all interrelated through the physical processes that connect the storage of water in
individual reservoirs and control the fluxes between reservoirs. Proper management of
this resource is essential for all societies. The potential for better management lies in
anticipatory and adaptive actions based on reliable medium and long term predictions and
projections regarding the availability and use of water. Research on the Global Water
41

DISCUSSION DRAFT
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35

Cycle is an essential contribution to the development of such predictive systems and


associated adaptive strategies.
Some of the challenges facing the global water cycle community will not be fully
resolved in this decade. Well thought out, stable and adequately funded programs are
needed to ensure that the activities will produce the needed results for the CCRI at the
same time as they are developing longer term objectives connected with the USGCRP.
This is possible because of the readiness of the science to take advantage of the new
observational systems that are coming on line and to utilize the new modeling capability
that is now becoming accessible to the water science community.
Water cycle research provides an excellent programmatic laboratory for exploring the
potential interactions between scientists and land and water managers who deal with
issues such as land, irrigation and fertilizer use and coastal zone protection. As this
aspect of the program grows, the priorities for water cycle research will be expanded. In
order to place the water cycle program in a position to deal with such growth issues, the
physical and natural sciences must be linked with social, legal and political issues.
In order to achieve a primary goals of the CCSP, namely reducing the uncertainties in the
global climate predictions/ projections, it is essential that the interactions between the
global water cycle and the climate system be more fully understood and be more
effectively simulated. The time is right to advance our understanding of many aspects of
the global water cycle. Some federal programs related to land surface forcing have
matured and are contributing many new insights regarding land surface process
understanding and ways to simulate these processes in models, while other programs are
receiving significant motivation and data sources through new technologies such as
advanced satellites sensors and data assimilation systems. Global water cycle issues need
to be addressed at a number of time and space scales. Hydrologic modeling is needed at
catchment and continental scales supplemented by a capability to scale up to regional and
continental scales. Atmospheric observational and modeling programs are needed to
address the role of the global water cycle in climate change. A holistic perspective of the
global water cycle needs to be developed through a strong Global Water Cycle program
so the entire system can be understood and the interconnectedness of its various
components can be properly addressed.

42

DISCUSSION DRAFT
1

References
2
3
4
5
6
7
8
9
10

Hornberger, G. M., J.D. Aber, J. Bahr, R. C. Bales, K. Beven, E. Foufoula-Georgiou, G.


Katul, J.L. Kinter III, R. D. Koster, D.P. Lettenmaier, D. McKnight, K. Mitchell,
J.O. Roads, B.R. Scanlon, and E. Smith. 2001. A Plan for A New Science
Initiative On Global Water Cycle. US Global Change Research Program,
Washington, D.C.
NRC, 1999. Our Common Journey: A transition toward sustainability. National
Academy Press. 363 pp.

43

Вам также может понравиться